Revising C

Nelson H.F. Beebe validgh!sun!magna.math.utah.edu!beebeaSun.COM
Tue Dec 11 16:46:52 PST 1990


Peter S. Shenkin
<uunet!avogadro.barnard.columbia.edu!shenkinauunet.UU.NET>
writes:

>> standards committee won't add features like noalias and
>> conformant arrays because there's been insufficient experience
>> with them, and no vendor wants to try them out because they're
>> not in the standard.
>> 
>> On the other hand, if these things could be tried out, and
>> turned out to be useful, there is at least the possiblity they
>> would be incorporated in a future standard.

I agree.

In the past, language progress happened slowly, because it took
large vendors years to make changes to their compilers.  Thanks
to the wonderful public services of the Free Software
Foundation, we now have a great test bed in their compilers for
proposed language extensions to be tried out before they are
considered for standardization.  The DoD Ada effort, I believe,
is an example of the trouble you can get into if you design an
language before you implement it.  The proposed Ada revisions
list is now quite a long document.

FSF's gcc compiler now produces code for more architectures
than probably any previous compiler, including pcc.  As it
evolves to support multiple language front ends, we move ever
closer to a test bed for new ideas that potentially can provide
very rapid implementation for a wide variety of architectures.

This, I trust, will provide the `prior art' that standards
committees hold dear, at the same time offering programmers the
possibility of immediate portability of their code to every
system that gcc runs on.  That too is important, because few
people today want to take on serious programming projects if
they are for a one-of-a-kind language or machine.


========================================================================
Nelson H.F. Beebe
Center for Scientific Computing
Department of Mathematics
220 South Physics Building
University of Utah
Salt Lake City, UT 84112
 Tel: (801) 581-5254
 FAX: (801) 581-4148
 Internet: beebeascience.utah.edu
========================================================================



More information about the Numeric-interest mailing list